Close
  Indian J Med Microbiol
 

Figure 1: Time-series forecasting work flow. The pipeline shows the workflow to accurately forecast the future values of a time-series with certain characteristics, e.g. specific trend. A set of candidates models are trained and the best one is selected based on the performance parameters, i.e., root mean squared error, determination coefficient (R2), 95% prediction interval, Akakie information criterion, and Bayesian information criterion

Figure 1: Time-series forecasting work flow. The pipeline shows the workflow to accurately forecast the future values of a time-series with certain characteristics, e.g. specific trend. A set of candidates models are trained and the best one is selected based on the performance parameters, i.e., root mean squared error, determination coefficient (R<sup>2</sup>), 95% prediction interval, Akakie information criterion, and Bayesian information criterion